Systems and methods for optimal privacy-preserving information revelation

Abstract

The present system relates a platform for addressing the optimal privacy-accuracy trade-off in the revelation of a user's valuable information to a third party. Specifically, the present system formalizes the privacy-accuracy trade-off in a precise mathematical framework, wherein mathematical formalization captures user's privacy preference with a single parameter. The system possesses a revelation method of user data that is optimal, in the sense of abiding by user's privacy preference while providing the most accurate description to third party subject to the aforementioned privacy preference constraint.

We claim: 
     
       1. A method of preserving privacy in a data set used to estimate information configured to be used by a third party, the method comprising the steps of:
 receiving an initial information data set and a user's privacy setting, the user's privacy setting including one or more privacy instructions defining conditions for sharing regarding the initial information data set; 
 using the initial information data set and the user's privacy setting as input, producing a privacy-preserving stochastic map using an updated prior knowledge data set, the user's privacy setting, and extraction noise statistics as inputs and then applying the privacy-preserving stochastic map to an extracted initial information data set to produce an adjusted information data set; and 
 using the adjusted information data set and the privacy-preserving stochastic map as inputs, applying a stochastic inference algorithm to produce an estimate of the initial information data set and an estimation error value, wherein the adjusted information data set is constrained to meet every condition defined within the user's privacy setting. 
 
     
     
       2. The method of  claim 1 , wherein the updated prior knowledge data set is a probability distribution over a set from which the initial information data set can take its values. 
     
     
       3. The method of  claim 1 , wherein the user's privacy setting includes a condition enabling the user to control a statistical distance between the adjusted information data set and the initial information data set. 
     
     
       4. The method of  claim 1 , wherein the stochastic inference algorithm minimizes an estimation error, wherein the estimation error is defined by an expected value of a distance between the initial information data set and the estimate of the initial information data set, where the distance is measured with respect to a given loss function. 
     
     
       5. The method of  claim 1 , wherein the privacy-preserving stochastic map minimizes the stochastic inference algorithm's estimation error subject to the user's privacy setting.